255. The capacity of the human eye for the perception of light is comparatively small. It is unable to perceive very faint lights, and it is dazzled and confused by lights of great brilliancy. Photographic plates are affected by faint lights which are invisible to the eye; thus, photographs of the sky reveal a multitude of stars which are not visible even with-.the aid of the strongest telescopes. The unaided eye is unable to judge of the relative intensity of various lights with any reasonable approach to accuracy.

256. The art of measuring the comparative intensity of lights is called photometry. There are several methods of making these measurements - chemical, electrical, and mechanical-each of which is peculiarly suited to special cases. The method employed for general purposes is to compare the illuminating power of the light under examination with that of a light of standard intensity.

The unit which is used for all ordinary measurements is the light given by a sperm candle, which burns at the rate of 120 grains per hour.

The candle is burned in still air, and care is taken to avoid all drafts which might accelerate the combustion, and thus vary the brilliancy of the light. The light thus obtained is made the unit for comparison, and is called one candlepower.

A larger unit is sometimes used for measuring very large lights. This is the flame of a certain variety of oil lamp called the Carcel lamp, and the unit thus derived is called one Carcel.

257. All instruments which serve to measure the comparative brilliancy of lights are properly called photometers.

One of the oldest of these instruments, called the Rum-ford photometer, is shown in Fig. 94. It consists of a table having a black wooden post c, standing erect as shown, and a screen g, which receives the shadows of the post that are cast by the lights a and b. The candle a is the standard light, and b is the light whose intensity is to be measured. The lines h c and i c make exactly equal angles with the screen, and the lights are moved back and forth along these lines until the shadows e and f appear of exactly equal blackness. The powers of the two lights are then computed by dividing the square of the distance b c by the square of the distance a c, the quotient being the candlepower of the light b.

Fig. 94.

This method is very inaccurate, and is not to be recommended, because the eye" is unable to compare the shadows e and f with the requisite accuracy.

258. The Bunsen photometer, shown in Fig. 95, operates upon a different principle. A diaphragm c is illuminated on its opposite sides by the light b and the standard candle a. The observer looks down through the tube e into mirrors f and g, and thus sees the reflection of both sides of the diaphragm at the same time. If they appear of unequal brilliancy, the sight-box d is moved along the bar h until they become equal. The candlepower of the light b is then found by dividing the square of the distance b c by the square of the distance a c; usually the bar is graduated, as shown, so that no calculation is necessary.

259. There are two methods of constructing the diaphragm in vogue. The spot diaphragm is shown in Fig. 96.

The center a is a disk of opaque white paper. The ring b is made of white paper which is saturated with paraffin, and is translucent. The outer part c is blackened. When this diaphragm is unequally illuminated on its opposite sides, the ring b looks darker, or brighter, than the center a, but when the illumination is exactly equal, all difference disappears, and the spot a becomes indistinguishable.

The star diaphragm is shown in elevation at A, and in section at B in Fig. 97. It consists of a piece of white writing paper a of moderate thickness, having a star-shaped figure cut out of its center, and a sheet of thin white writing paper c of best quality, which is doubled so as to enclose the piece a. The diaphragm is lightly squeezed between two pieces of glass b, b. Care is taken in cutting the star to make every point and line clear and sharp. When the reflection of the diaphragm is seen in the mirrors, the images will vary in distinctness if the lights are unequal. The sight-box d in Fig. 95 is then moved along the bar until both images of the star appear equally sharp and clear.

It will be observed that the methods of testing employed in the photometers described are quite different. In the Rumford method the observer judges the equality in blackness of the shadows produced; in the Bunsen method, using the spot diaphragm, he judges by the equal brightness of the opposite sides of the diaphragm, and when using the star diaphragm, he judges by the equal clearness and distinctness of the two images of the star.

Fig. 96.

Fig. 97.

The Rumford method has been discarded for the more accurate Bunsen method. Both the spot and star diaphragms are widely used; but the star diaphragm is preferred because of its superior accuracy.

260. In practice the distance between the centers of the lights, Fig. 95, is usually made 100 inches, and the bar is graduated according to the following table, the numbers given being the distance in inches from the center of the candle flame to the center of the diaphragm for each candle-power.